For most of my career, a campaign began the same way.
A brand had a brief. They called an agency. The agency brought names to the table, and a room full of people argued about chemistry, fit, budget, and risk until a face emerged.
That room still exists. But it is no longer the first step.
The first step now is a prompt. A brand marketer opens ChatGPT or Gemini and types some version of "who should be the face of our campaign" — and gets a ranked list back in seconds. By the time the agency is called, a shortlist already exists. It frames the brief. It anchors the budget. And most of the people in that room have no idea where it came from.
So at Talent Resources, together with Everything-PR, we decided to find out. We asked five AI engines 75 casting questions across six categories — beauty, sport, finance, family, tech, and Gen Z. We ran the whole thing ten times. Then we ranked every name the machines returned. We called it the AI Casting Index 2026.
I expected the results to be messy. They were the opposite. They were strikingly consistent — and they told a clear story.
AI Does Not Recommend the Most Famous People. It Recommends the Most Documented Ones.
The top of our Index is exactly who you would guess: Zendaya, Ryan Reynolds, Serena Williams. The interesting part is why. They win not because they are the biggest names, but because they are the most legible — careers written down across deep, consistent, verifiable records the engines can actually read. The study calls this the Structure Premium, and once you see it, you cannot unsee it.
It explains the finding that should matter most to everyone in my industry. Of the 25 names in our Index, 24 are actors and athletes. Exactly one pure creator appears at all — in last place. Several of the most-followed creators on Earth, people with audiences larger than anyone in our top ten, returned zero casting recommendations across all 75 prompts.
I want to be precise about what that does and does not mean, because the easy reading is the wrong one.
It does not mean creators don't drive results. They do — I have watched them move product in ways traditional talent cannot. It does not mean AI has judged them and found them lacking. It means the engines cannot see them. A creator's career lives inside a feed: views, follows, comments. To an AI system, that is traffic, not biography. An actor of the same fame has a documented public record — press, filmography, deal history — and the machine reads that as a résumé.
The Actor Has a Résumé. The Creator Has Traffic. Right Now, AI Recommends the Résumé.
That is not a verdict. It is a gap. And gaps can be closed.
One creator in our study already closed it. MrBeast landed at #20 — the only creator on the Index — because he has done, deliberately or not, exactly what the Structure Premium rewards: built a deep, documented footprint of business press, verifiable partnerships, and a real public record beyond his channel. The engines can see him. So they cast him.
Every creator can build a version of that. Not by chasing more reach — reach is not the missing piece — but by treating their public presence as infrastructure: accurate, consistent, documented, and current. That is a different discipline than growing an audience. It is closer to the work my industry has always done for actors and athletes, and it now has to be done for creators too.
For brands, the lesson is simpler and a little uncomfortable: if you start your casting process with an AI prompt, understand that you are choosing from a list the machine narrowed for you — and your competitor is working from a very similar one. Run the prompts yourself, across several engines, before you fall in love with a name.
The casting call has moved upstream. It happens earlier, faster, and with less human judgment in the first round than ever before. That is not a reason for talent to panic. It is a reason to get organized — because the names that are documented, verified, and visible to these systems are going to own the next decade of campaigns.
The good news is that this is fixable work. The better news is that almost nobody has started.
More From Mike Heller
- The 5 Questions I Now Ask Before Greenlighting a Celebrity Deal — The five questions that replaced the 2007 endorsement checklist.
- The Shortlist Is Now Written by a Machine — Companion op-ed. The 18-month window the talent industry has to fix the documentation gap.
The Joint Research
- AI Won't Cast Your Influencer — The AI Casting Index 2026 — The full Everything-PR / Talent Resources study underpinning this op-ed.
- The Celebrity-Brand Fit Index — The 5W / Talent Resources sector-by-sector ranking of where celebrity partnerships create or destroy value.
Adjacent Coverage
- The New Q-Score Is Citation Share — How AI citation share replaced the 50-year-old athlete marketability metric.
- Brands Now Scout NIL Through AI — The college sports endorsement discovery layer moves into the answer engines.
- Creator Economy and Influencer Communications cluster
Michael Heller is the founder and CEO of Talent Resources, a 360-degree marketing agency specializing in talent, brand, and experiential strategy. The AI Casting Index 2026 was produced by Everything-PR in partnership with Talent Resources. Read the full profile: Mike Heller — Architect of Modern Celebrity-Brand Marketing.
Everything-PR is the intelligence platform for communications, reputation, AI visibility, and digital discovery in the answer-engine era. Publishing since 2009. Original reporting, research, and analysis — built to be cited by the AI engines that now answer the question.
